def test__crop_img_to():
data = np.zeros((5, 6, 7))
data[2:4, 1:5, 3:6] = 1
affine = np.diag((4, 3, 2, 1))
img = nibabel.Nifti1Image(data, affine=affine)
slices = [slice(2, 4), slice(1, 5), slice(3, 6)]
cropped_img = image._crop_img_to(img, slices, copy=False)
new_origin = np.array((4, 3, 2)) * np.array((2, 1, 3))
# check that correct part was extracted:
assert_true((cropped_img.get_data() == 1).all())
assert_true(cropped_img.shape == (2, 4, 3))
# check that affine was adjusted correctly
assert_true((cropped_img.get_affine()[:3, 3] == new_origin).all())
# check that data was really not copied
data[2:4, 1:5, 3:6] = 2
assert_true((cropped_img.get_data() == 2).all())
# check that copying works
copied_cropped_img = image._crop_img_to(img, slices)
data[2:4, 1:5, 3:6] = 1
assert_true((copied_cropped_img.get_data() == 2).all())
def _generate_preprocessed_samples_core(path_brats_preprocessed,
sample,
mean_samples,
std_smaples,
resized_img_shape=(144, 144, 144)):
modality_images = []
modality_filenames = []
for mod in modalities[:-1]:
mod = '*' + mod + '.nii.gz'
mod_wise_files = glob(os.path.join(sample, mod))
assert len(mod_wise_files) == 1, 'Minimum one modality nifti image expected'
modality_filenames.append(mod_wise_files[0])
slices = _crop_to_non_zero_content(modality_filenames)
for mod_fn in modality_filenames:
data = nib.load(mod_fn)
data = _crop_img_to(data, slices, copy=True)
data, new_spacing_mi = _resize_nifti_images(data, interpolation='linear', resized_img_shape=resized_img_shape)
modality_images.append(data)
modality_images = np.asarray(modality_images, dtype=np.float32)
modality_images -= mean_samples[:, np.newaxis, np.newaxis, np.newaxis]
modality_images /= std_smaples[:, np.newaxis, np.newaxis, np.newaxis]
affine = np.eye(4)
new_spacing_mi = np.append(new_spacing_mi, 1)
np.fill_diagonal(a=affine, val=new_spacing_mi)
mod_data_resized_nifti = nib.Nifti1Image(modality_images, affine=affine)
mod_filename = os.path.basename(sample) + '_mod.nii.gz'
nib.save(mod_data_resized_nifti, os.path.join(path_brats_preprocessed, mod_filename))
seg = '*' + 'seg' + '.nii.gz'
seg_file = glob(os.path.join(sample, seg))
assert len(seg_file) == 1, 'Minimum one segmeantion image expected'
seg_data = nib.load(seg_file[0])
seg_data = _crop_img_to(seg_data, slices, copy=True)
seg_data, new_spacing_seg = _resize_nifti_images(seg_data, interpolation='nearest', resized_img_shape=resized_img_shape)
seg_data = seg_data[np.newaxis]
affine = np.eye(4)
new_spacing_seg = np.append(new_spacing_seg, 1)
np.fill_diagonal(a=affine, val=new_spacing_seg)
seg_data_resized_nifti = nib.Nifti1Image(seg_data, affine=affine)
seg_filename = os.path.basename(sample) + '_seg.nii.gz'
nib.save(seg_data_resized_nifti, os.path.join(path_brats_preprocessed, seg_filename))
def _generate_mean_std_from_sample(sample, resized_img_shape):
modality_images = []
modality_filenames = []
for mod in modalities[:-1]:
mod = '*' + mod + '.nii.gz'
mod_wise_files = glob(os.path.join(sample, mod))
assert len(mod_wise_files) == 1, 'Minimum one modality nifti image expected'
modality_filenames.append(mod_wise_files[0])
slices = _crop_to_non_zero_content(modality_filenames)
for mod_fn in modality_filenames:
data = nib.load(mod_fn)
data = _crop_img_to(data, slices, copy=True)
data, _ = _resize_nifti_images(data, interpolation='linear', resized_img_shape=resized_img_shape)
modality_images.append(data)
modality_images = np.asarray(modality_images)
std = np.std(modality_images, axis=(-3, -2, -1))
mean = np.mean(modality_images, axis=(-3, -2, -1))
return mean, std
